Back to skill
Skillv1.0.0
ClawScan security
Nano Banana 2 — AI Image Generation (Gemini 3.1 Flash Image, Google, Evolink) · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 26, 2026, 5:12 PM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requirements and runtime instructions are coherent with an Evolink image-generation integration and do not request unrelated credentials or installs.
- Guidance
- This skill appears internally consistent: it only needs an Evolink API key and uses Evolink domains for uploads and generation. Before installing, verify you trust Evolink.ai (check the vendor/site and terms), use a dedicated API key with limited scope if possible, monitor billing/usage (requests can consume credits), and rotate/revoke the key if you stop using the skill. Note the registry lists no source/homepage entry — consider confirming the publisher identity (e.g., GitHub/npm links mentioned in SKILL.md) before providing your API key. Autonomous invocation is allowed by default; if you want to prevent background calls, disable autonomous invocation in your agent settings.
Review Dimensions
- Purpose & Capability
- okName/description advertise Evolink + Gemini image generation and the skill only requires EVOLINK_API_KEY and documents Evolink endpoints and model IDs — these align with the stated purpose.
- Instruction Scope
- okSKILL.md contains only workflow steps for uploading files, calling generate_image, polling tasks, and basic UX guidance; it does not instruct reading unrelated system files or exfiltrating data to unknown endpoints.
- Install Mechanism
- okThere is no install spec and no code files to execute; this instruction-only skill relies on existing MCP tooling or simple HTTP calls, which minimizes on-disk risk.
- Credentials
- okOnly one env var (EVOLINK_API_KEY) is required and it directly maps to the declared API usage; there are no unrelated secrets or config paths requested.
- Persistence & Privilege
- okThe skill is not always-enabled and does not request system-wide modifications or persistent privileges. Autonomous invocation is allowed (default) but this is typical for skills and not combined with other risky behavior.